PEER REVIEWED

An Update on Primary Malignant Intraocular Tumors

AMY C. SCHEFLER, MD · MARIA ELENA JOCKOVICH, PhD · IZIDORE S. LOSSOS, MD · STUART TOLEDANO, MD · MICHAEL FEILMEIER, MD · SANDER R. DUBOVY, MD · TIMOTHY G. MURRAY, MD, MBA

RETINOBLASTOMA

Classification

The staging of retinoblastoma is currently an area of controversy among clinical centers. The Reese-Ellsworth Classification for intraocular retinoblastoma, developed in the 1960s, has been used almost universally since its inception. It was originally developed to predict prognosis in eyes treated with lateral port external-beam radiation. In recent years, its usefulness has been questioned by some centers because external-beam radiation is used less frequently and a new international classification scheme has been developed (Table 1). The new classification has been used clinically in several centers in a collaborative study.¹

The subgroups in the new classification scheme range from the least advanced intraocular disease, most easily treated with chemotherapy and focal therapy, to the most advanced disease, least easily cured with these methods. At the 1 center in which outcomes have been retrospectively correlated with the new classification scheme, the classification was shown to predict outcome from this treatment approach in a stepwise fashion with decreasing globe salvage for higher-staged eyes.² In this study, globe salvage rates of 100% were achieved for group A, 93% for group B, 90% for group C, and 47% for group D cases. Group E patients were primarily managed with enucleation.

Multiple problems persist in the application of the new classification. Comparisons with older series will be difficult if Reese-Ellsworth groups are not also included in newer research. Most likely, the majority of authors will continue to include the Reese-Ellsworth classifications for an indefinite transition period for ease of comparison. Second, because this scheme predicts globe salvage rates based on 1 current treatment modality, it could become obsolete in the same manner as the Reese-Ellsworth classification if treatment trends shift in the future.³

Systemic Chemotherapeutic Agents and Focal Therapies

Chemotherapy has been used in the treatment of retinoblastoma since the 1950s, but its widespread use as a standard of care was not established until the mid-1990s. By then, external-beam radiation had begun to fall out of favor because research had demonstrated its potential to increase the risk of developing second cancers in survivors of germinal retinoblastoma.⁴ As a result, many clinicians became interested in the use of chemotherapeutic regimens for intraocular retinoblastoma that could be successful while eliminating the need for external-beam radiation.

There is no universal agreement about the ideal chemotherapeutic regimen. The number of agents and dosing schedules continue to vary greatly among centers. The most common regimen used in ocular oncology centers since 1996 has been a 3-drug combination of vincristine (typically 0.05 mg/kg for children 36 months or younger and 1.5 mg/m² for older children), carboplatin (18.6 mg/kg for children 36 months or younger and 560 to 600 mg/m² for older children), and etoposide (5 mg/kg for children 36 months or younger and 150 mg/m² for older children). However, the number of doses and number of cycles of chemotherapy, as well as the approach to focal therapy, have varied greatly at major centers with no 2 sites treating these patients in the same manner.^5-25 Even in single centers, the treatment approach has evolved over time. Some centers have used a 1- or 2-drug combination regimen to reduce systemic toxicity and avoid a chemotherapy-induced second malignancy (for more information: www.retinalphysician.com/table1.htm).^{9,12,13,15,18,21,25} Other investigators have supported the addition of cyclosporine as a P-glycoprotein inhibitor to decrease the ability of tumor cells to transport antineoplastic drugs from the intracellular space, allowing the cells to develop multidrug resistance.^7,26 Most centers now use chemotherapy to shrink tumors so they are then amenable to focal treatments such as laser or cryotherapy. The effect on globe salvage rates of the temporal relationship between the application of chemotherapy and focal therapy is unknown. In our center, focal treatments are applied the same day that chemotherapy is administered in an attempt to achieve a synergistic effect, an approach supported by clinical and laboratory data (Figure 1).^22,27

Figure 1. (top) Large retinoblastoma tumor before treatment. (bottom) After treatment with systemic chemotherapy (vincristine, carbopatin, etoposide) and aggressive focal laser therapy.

We recently published an analysis of our series of retinoblastoma patients treated with 4 to 9 cycles of 3-drug chemotherapy.²⁴ Our analysis revealed that a 3-drug regimen of carboplatin, vincristine, and etoposide, when combined with aggressive repetitive application of focal diode laser, results in superior globe salvage rates in disease stages when compared with regimens of fewer drugs, fewer cycles of chemotherapy, or less aggressive application of laser (Tables 2A and 2B). Moreover, all the patients in this cohort had retinoblastoma presenting in the macula, which tends to present in younger children and demonstrate a more aggressive clinical course than peripheral tumors that present later in life.²⁸ Patients in our cohort were treated under anesthesia with diode laser at every examination until their tumors were noted to be inactive for at least 6 months. The 1.2-mm spot size was aimed through a dilated pupil using the indirect ophthalmoscope and a 20 D lens. The laser was typically set at 350 mW initially; then the power was increased as necessary until a gray-white appearance of the tumor was achieved. The entire surface of the tumor was treated with areas of overlap to ensure that no portion of tumor was missed. Our patients had a rate of avoidance of external-beam radiation and enucleation at 3 years in 100% of patients with Reese-Ellsworth group I-IV disease and 83% in patients with Reese-Ellsworth group V disease. These tumor-control rates far exceed those published elsewhere. No prospective randomized studies with extended follow-up comparing different chemotherapeutic regimens or focal therapy protocols have been performed to date, but such a trial is essential in the future in order to establish the ideal treatment method and clinical standardization.

Periocular/Subtenon Carboplatin Use

During the mid-1990s, our group and others began to investigate local chemotherapeutic drug delivery to the eye in animals.^29,30 Once the lack of toxicity and effectiveness of drug delivery to the vitreous were demonstrated in the animal model, several studies evaluating the efficacy and toxicity of periocular carboplatin for human retinoblastoma followed.^31,32 Currently, most large centers utilize this technique in conjunction with systemic chemotherapy for salvage of advanced disease that is refractory to other forms of treatment. However, no standard indications or protocol for the application of subtenon carboplatin administration exists. One of the Childrens' Oncology Group (COG) studies currently enrolling patients will address this clinical question.

New Pharmacotherapies for Retinoblastoma

The limited systemic toxicity experienced by humans undergoing periocular chemotherapy has provided a new rationale for investigating novel ocular drugs that previously would be unacceptably toxic if administered intravenously. Angiogenesis has been shown to be essential for tumor survival for every cancer type, and there has been a recent explosion of interest in the use of antiangiogenic agents for cancer therapy. Furthermore, there has been increasing interest in the administration of antiangiogenic compounds in the eye for various ocular diseases in which neovascularization and increased vascular permeability occur, most notably in age-related macular degeneration (AMD). Retinoblastoma is highly vascularized and dependent on the ongoing development of a neovascular bed. Retinoblastoma forms cuffs of cells that surround blood vessels, and areas of necrosis are found removed from vessels.³³ This dependence on vascular supply, along with the anatomically isolated intraocular cavity, makes retinoblastoma a potential ideal target for antiangiogenic therapy.³⁴

Our group first reported on the use of an antiangiogenic compound for retinoblastoma by observing the effect of subconjunctival injections of combretastatin A-4 phosphate (CA-4P) prodrug treatment on tumor vasculature and growth in the luteinizing hormone ß (LHß)-Tag transgenic mouse.³⁵ The CA-4P prodrug induced an extensive, dose-dependent decrease in microvessel density and led to significant tumor reduction in treated eyes compared with the placebo control (P<.001). No evidence of intraocular toxicity was observed by histopathologic evaluation.

Our group also recently reported a decreased tumor burden in LHß-Tag mice compared to subconjunctival carboplatin injections alone when these injections were combined with subconjunctival injections (150 to 300 g) of anecortave acetate, an angiostatic cortisone (modified steroid).³⁶ The greatest reduction in tumor burden was observed when a low-dose (150 to 300 g) injection of anecortave acetate was given after 6 subconjunctival carboplatin injections. No retinal toxicity was observed.

Our laboratory has recently explored the potential use of antiangiogenic agents for human retinoblastoma by characterizing tumor vessel maturation patterns in the LHß -Tag mouse.³⁷ Immunofluorescence studies indicate that increased cell proliferation and angiogenesis are detected in the retinal inner nuclear layer before morphologic neoplastic changes occur. As tumor size increases, angiogenesis diminishes concomitantly with the appearance of mature vessels. Treatment with antiangiogenic drugs reduces the amount of immature vessels, but not mature vessels in this mouse model. This study suggests that patients with small, new tumors could be treated with an antiangiogenic agent such as anecortave acetate, while advanced tumors may require adjuvant treatments targeting mature vessels.

Intra-arterial Administration

Clinicians in Japan first postulated that another method of local drug delivery, intra-arterial chemotherapy infusions, may increase penetration into small intraocular tumors and decrease systemic side effects.³⁸ In a study of 187 patients, melphalan (Alkeran, Celgene) infusions were performed using a catheter tip placed just distal to the orifice of the ophthalmic artery with temporary occlusion of the internal carotid artery during the infusion. Bradycardia during injection, facial erythema, and mild eyelid edema were common side effects. Local tumor-control rates and eye-preservation rates were not reported. Abramson and colleagues³⁹ have begun treating patients with advanced retinoblastoma with similar selective ophthalmic artery infusions of melphalan. No local ocular, orbital, or motility-related adverse effects, systemic toxicity, or infections were reported. Visual acuity (VA) results have not been published.

COG Clinical Trials

The COG has initiated several National Cancer Institute-funded multicenter prospective clinical trials to address some of the controversial issues in the care of retinoblastoma patients (for more information: www.retinalphysician.com/table1.htm). Three trials are currently in progress and are actively recruiting patients. The first trial is a phase 3 single-arm trial for systemic and subtenon chemotherapy for International Classification groups C and D (advanced) retinoblastoma. The rationale for this study is that, while many centers around the country utilize subtenon carboplatin for patients with advanced disease, there is no standard regarding frequency or timing of dosing. Furthermore, the study aims to verify the results of chemotherapy trials in single institution reports and to achieve a higher salvage rate of functional eyes while limiting the use of external beam radiation and its side effects.

Patients in this trial will receive systemic intravenous vincristine, etoposide, and carboplatin. They will receive subtenon carboplatin on day 0 of courses 2 through 4 only. Treatment will repeat every 28 days for 6 courses, barring occurrence of extraocular retinoblastoma or second malignancy. Beginning with course 3 of systemic chemotherapy, patients will undergo focal therapy with laser and/or cryotherapy on cycle day 1. The primary outcome measure is event-free survival at 12 months; secondary outcome measures include acute and long-term toxic effects, failure patterns, predictors of failure, and percentage of preservation without enucleation after failed treatment.

The second COG trial will examine neoadjuvant chemotherapy (vincristine and carboplatin) for International Classification group B retinoblastoma. The purpose of this study is to investigate the outcomes of a 2-drug regimen (to minimize systemic toxicity and the risk of etoposide-induced second cancers) in patients with less-advanced retinoblastoma. Previous single-institution work has not demonstrated whether eliminating etoposide from the chemotherapeutic regimen raises the risk of tumor recurrence. Patients will receive systemic intravenous carboplatin and vincristine. Treatment will repeat monthly for 6 cycles barring disease progression or unacceptable toxicity. Patients will undergo focal therapy with laser, cryotherapy, or brachytherapy at the discretion of the ophthalmologist after the first chemotherapy cycle. The primary outcome measure is 2-year event-free survival (need for additional chemotherapy, external beam radiation, or enucleation).

The third trial will examine the role of adjuvant chemotherapy in patients with unilateral retinoblastoma who have undergone enucleation with and without high-risk pathologic features. This study's purpose is to investigate what adjuvant chemotherapy is necessary to prevent metastatic disease. Previous single-institution studies have used a wide variety of indications for chemotherapy in patients and a variety of drugs, and so the approach to adjuvant therapy is not standardized across institutions. The study will enroll patients who have just undergone enucleation for unilateral retinoblastoma. Patients with high-risk histopathologic features (defined in this study as massive choroidal involvement, any uveal involvement with any degree of optic nerve invasion, and isolated optic nerve involvement posterior to the lamina cribrosa) will receive 6 cycles of systemic vincristine, carboplatin, and etoposide. The primary outcome measures of this study are event-free (defined as the avoidance of extraocular or metastatic disease) and overall survival at 2 years and the secondary outcome measure is toxicity.

Diagnostic Imaging for Retinoblastoma

One potential challenge for the future development of adjuvant pharmacotherapies such as anecortave acetate is the successful delivery of the drug. Our group has investigated the use of a blunt cannula delivery system to inject anecortave acetate reliably in a juxtascleral posterior subtenon location in animals.⁴⁰ Ultrasonography and magnetic resonance imaging confirmed the location of the drug during and after injection. The drug remained localized to the injection site up to 5 weeks after initial injection. Furthermore, along with other investigators at our institution, we have developed a high-resolution spectral-domain OCT system for in vivo imaging of the rodent retina.⁴¹ This system will be highly useful in the future for monitoring the response to experimental therapies in the mouse.

CHOROIDAL MELANOMA

Genetic Analysis

Several chromosomal abnormalities, including gain or loss of chromosomal material in chromosomes 3, 6, and 8, have been detected in primary uveal melanoma tissue and have been associated with metastasis.⁴² Monosomy 3 in uveal melanoma has been shown to be a statistically significant predictor of both relapse-free and overall survival.⁴³ Prescher and colleagues reported that, while 57% of patients with monosomy 3 developed metastases within 3 years, no patients with disomy 3 developed metastatic disease.

Global gene expression patterns have been examined by gene expression microarray analysis of 3075 genes in 25 enucleated eyes.⁴⁴ Two groups were identified: class 1 (low-grade tumors) and class 2 (high-grade tumors). Class 2 tumors demonstrated downregulated gene clusters on chromosome 3 and upregulated clusters on chromosome 8q. These classifications strongly predicted metastatic death with a 95% Kaplan–Meier-based survival prediction at 92 months of 95% in class 1 and 31% in class 2. The classifications outperformed other pathologic prognostic indicators.

Recently, the number of clinical centers performing karyotyping, single nucleotide polymorphism analysis, fluorescent in situ hybridization (FISH) analysis, and/or comparative genomic hybridization on fine needle aspiration and enucleation specimens has increased dramatically. Some groups have also begun to perform FISH or microsatellite array on fine needle aspiration biopsy specimens.^45,46 Biopsies have been attempted via pars plana and trans-scleral approaches. In recent series, sufficient tissue material for diagnosis was obtained for FISH and microsatellite assay in 98% and 86% of cases, respectively.^45,46 As no effective treatment is available for metastatic uveal melanoma at this time, it is unclear what interventions or screening programs should be offered to patients whose lesions are shown to be high-risk for metastasis on biopsy. Nonetheless, as new treatments or prophylactic drugs for metastatic melanoma are developed and available for clinical trials, it will be important to identify which patients are appropriate for entry into such studies.

Treatment of Small Melanomas

Since the Collaborative Ocular Melanoma Study (COMS) medium tumor trial was published, most ophthalmologists in the United States have continued to use the COMS tumor-size guidelines in order to determine whether brachytherapy is an appropriate treatment for a particular patient. Many investigators have raised the question of whether small melanomas should also be treated with brachytherapy. Although the COMS observational study of small melanomas indicated that the long-term mortality rate for these patients is low (8-year all-cause mortality of 14.9%),⁴⁷ it may be significant enough to justify the consideration of brachytherapy in these patients.⁴⁸

Only 1 study has examining the outcomes of patients with small melanomas treated with standardized brachytherapy. Sobrin and colleagues⁴⁹ reported a 3.9% (95% confidence interval, 0% to 11.2%) melanoma-specific 5-year mortality rate in patients with small melanomas who were observed and then treated with iodine-125 brachytherapy at the time of either observed growth or the development of orange pigment. The only other study examining this group of patients reported a 5-year melanoma-specific mortality rate of 5.8%.⁵⁰ The melanoma-specific 5-year mortality rate for the tumors in the COMS small-tumor study was 1%,⁴⁷ but the calculation of this rate included a substantial number of patients whose suspected tumors did not grow and were never treated and are therefore at lower metastatic risk than the patients in the Sobrin et al. and Butler et al. studies. Moreover, there was no defined criteria or threshold for treatment in the COMS small-tumor trial and this group may have included patients whose lesions demonstrated no growth or high-risk features. The treated group (when considered alone) in the COMS thus must have had a higher death rate, but this was not calculated in the report. In the Sobrin et al. study, the overall mortality of all patients with small lesions (including those that were observed) was 0.7%, proving that initial observation did not result in a higher mortality.

Figure 2. (top) Amelanotic choroidal melanoma after plaque therapy with radiation retinopathy. Note macular exudate and hemorrhage along arcades. (middle) OCT demonstrating severe macular edema due to radiation maculopathy. (bottom) OCT 6 months after injection of intravitreal Avastin.

A randomized, prospective trial of visual and survival outcomes in patients managed by observation vs prompt treatment is needed to answer this critical question.⁴⁹ Singh and colleagues⁵¹ have proposed such a trial that will enroll patients with small tumors (size by COMS criteria) and patients will be randomized to immediate treatment vs deferred treatment (observation with growth prior to treatment). In order to decrease the number of patients needed for sufficient power to detect a mortality difference in this study, only patients whose lesions demonstrate risk factors for growth (height >2 mm, orange pigment, subretinal fluid) will be included. Five years of follow-up is planned.

New Treatments for Radiation Retinopathy

A common late postoperative complication of brachytherapy in choroidal melanoma patients is radiation retinopathy (RR). RR is a slowly progressive, occlusive vasculopathy characterized by radiation-induced endothelial damage. Kaplan-Meier analysis has demonstrated rates of nonproliferative and proliferative RR at 5 years of 42% and 8%, respectively.⁵² Risk factors for the development of nonproliferative RR include tumor margin of less than 4 mm from the fovea and radiation dose rate of greater than 260 centigrays/hour to the tumor base.⁵² Risk factors for the development of proliferative RR include diabetes and tumor base greater than 10 mm.⁵²

Radiation retinopathy may be treated with panretinal photocoagulation (PRP) when proliferative disease is present.⁵³ A small study reviewed the effect of performing prophylactic PRP prior to the development of clinically evident RR. None of the patients lost <3 lines of vision.⁵⁴ Macular edema can be managed with focal grid laser and intravitreal triamcinolone acetonide, although the effect of these treatments is often temporary.⁵⁵ Recently investigators have begun experimenting with intravitreal injections of anti-vascular endothelial growth factor (VEGF) compounds such as bevacizumab (Avastin, Genentech) to treat radiation retinopathy in patients treated with iodine 125,^56,57 palladium 103,⁵⁸ and ruthenium 106.⁵⁹ These studies contained very few patients and short follow-up. Our unpublished experience with over 100 patients indicates that bevacizumab is very effective in achieving short-term resolution of macular edema due to RR, but long-term data on this treatment are not yet available (Figure 2).

Ultimately, end-stage RR can result in neovascular glaucoma. These patients can be treated with topical therapy.⁶⁰ Cyclophotocoagulation or retrobulbar alcohol injection can be performed for patient comfort if preservation of the globe is desired. Ultimately, however, many blind, painful eyes require enucleation after severe RR develops. Fortunately, this outcome is rare and occurs in less than 1% of patients undergoing brachytherapy at our institution.

Diagnostic Imaging for Choroidal Melanoma

As noted earlier, 1 of the clinical challenges in caring for patients with small choroidal melanomas is predicting which lesions will demonstrate growth. Some recent work has focused on the use of optical coherence tomography (OCT) to detect small amounts of subretinal fluid adjacent to small lesions. Since subretinal fluid is a known risk factor for growth,⁶¹ OCT may become a more integral modality in the routine clinical follow-up of these patients.

Autofluorescence patterns of small melanomas have also been recently described. Most choroidal melanomas appear to have a pattern of confluent hyperfluoresence, whereas most nevi do not.⁶² Autofluorescence may be a noninvasive tool to assess lipofuscin in these lesions, another known risk factor for growth and malignant transformation.⁶³

LYMPHOMA

Primary intraocular lymphoma (PIOL) is a rare malignancy, estimated to represent less than 1% of non-Hodgkin's lymphomas and less than 1% of intraocular tumors.⁶⁴ Clinically, the disease is considered a subcategory of primary central nervous system (CNS) lymphoma,⁶⁵ but recent molecular studies suggest the existence of differences in disease pathogenesis.⁶⁶ It is hypothesized that PIOL originates from lymphocytes that are chronically activated, selected, and transformed by exposure to a pathogen or antigen in the eye. Alternatively, the lymphocyte transformation might occur outside the eye and the malignant cells may become trapped in the "immune-privileged" structures of the eye that permit tumor growth while the systemic clone is eliminated by the intact immune system. The disease is characterized by malignant lymphocytic invasion of the vitreous, retina, or optic nerve, and it may occur either as an isolated finding before involvement of the neuraxis is apparent or as a component of concomitantly or previously diagnosed central nervous system disease.

The incidence of primary CNS lymphoma has increased over the past 30 years, in part due to the increased incidence of AIDS.⁶⁷ Since the mid-1990s, however, the incidence appears to have stabilized with a decrease in young patients and those with AIDS due to the availability of effective antiretroviral therapies.⁶⁸ There continues to be an increase in incidence in the older immunocompetent population that is not accounted for by improved neuroimaging or stereotatic neurosurgery.⁶⁷

Diagnostic Approaches

Patients with PIOL may manifest a variety of ophthalmic signs and can easily mimic other intraocular conditions. In a review of over 800 uveitis patients in a large referral center, this diagnosis represented the most common masquerade syndrome, accounting for one-third of all masquerade syndromes although it only represented 1.6% of all uveitis.⁶⁹ Patients can present with anterior-chamber inflammation, hyphema, hypopyon, iris neovascularization, iris or angle mass, vitritis, vitreous hemorrhage, subretinal or subretinal pigment epithelial infiltrates, retinal hemorrhage or exudate, perivascular infiltrate, retinitis, or optic-disc edema (Figure 3). As a result, they frequently initially present to a retina specialist.

Definitive diagnosis of primary intraocular lymphoma can only be made with either fluid or tissue analysis. Pars plana vitrectomy (PPV) is the most commonly used and most effective sampling technique and can provide both diagnostic and therapeutic results. The advantage of PPV is that the surgeon can obtain a large-volume fluid sample and targeted tissue biopsy facilitating cytological analysis.⁶⁵ Recently, the advent of the 23-g PPV system has enabled a sutureless procedure while maximizing the specimen yield.

Particularly in diagnostically challenging cases, we typically send 2 types of specimens. The first is a minimally diluted vitreous in order to minimize mechanical trauma to tumor cells. This specimen is collected by gentle manual aspiration of a 3-cc syringe that is attached to the stopcock adjacent to the aspiration line. Since the lymphoma cells tend to undergo spontaneous lysis, the sample needs to be delivered immediately to the pathology laboratory. The sample is then sent for flow cytometry and gene rearrangement studies. The second specimen is the diluted specimen in the vitrector cassette which is used for cytologic analysis.

Figure 3. (top) A patient presenting with bilateral primary intraocular lymphoma. The patient's vision was 20/50 in the right eye and 20/60- in the left eye. (bottom) The patient's vision recovered to 20/20 in both eyes after treatment with external beam radiation therapy and methotrexate-based chemotherapy.

Because diagnostic vitrectomies are often negative, patients require multiple procedures before a definitive diagnosis is identified. Recently, French researchers reported that the measurement of interleukin-10 levels in aqueous can serve as a screening tool for primary intraocular lymphoma.⁷⁰ A cutoff of 50 pg/mL in the aqueous was associated with a sensitivity of 0.89 and specificity of 0.93.

Current Treatments

Primary intraocular lymphoma is challenging to treat and requires close interaction between the retina specialist and ocular oncologist. The concentrations of drug delivered to the vitreous via intravenous injection are low and unpredictable, requiring very high dosing of the chemotherapeutic agents. Traditionally the mainstay of treatment was external beam radiation with a 3000- to 4500-centigray treatment prescribed to both eyes with historical 5-year survival rates dismally low at 10% to 29%.⁶⁷ Furthermore, morbidity due to neurotoxicity is common, especially in older patients, and can present as cognitive dysfunction, ataxia, and dementia.⁷¹ Better disease control combining radiation and methotrexate-based chemotherapy has been reported,⁷² but the combined strategy is not universally accepted based on retrospective surveys of routine clinical practice.⁶⁷ Newer therapeutic strategies have included high-dose chemotherapy with autologous stem-cell transplantation⁷³ and intravitreal injections of methotrexate with or without thioTEPA.^74-76 In the largest of these studies, Smith and colleagues⁷⁶ treated 16 patients with a series of intravitreal methotrexate injections performed twice weekly for a month, then weekly for a month, and then monthly for a year. Short-term remission (median of 18.5 months of follow-up) was achieved in all patients who completed the protocol, although 6 patients died of progression of intracranial disease. A confounding factor in this study was that patients in the study received a wide range of additional treatments, including chemotherapy of varying regimens and ocular external beam radiation. This heterogeneity makes it difficult to assess the efficacy of the intravitreal methotrexate in isolation.

Update on Investigational Treatments

An investigational strategy that has gained attention is the use of intravitreal rituximab. Rituximab (Rituxan, Genentech) is a chimeric monoclonal antibody directed against the B–cell-specific antigen CD20 and is efficient alone or in combination with chemotherapy in systemic non-Hodgkin's lymphoma.^67,71 High doses can be safely administered intravenously to achieve higher concentrations in the cerebrospinal fluid.⁶⁷ Early promising results in primary central nervous system lymphoma have been reported in a few cases of leptomeningeal lymphoma treated by intraventricular administration⁷⁷ and when used as salvage therapy by itself or in combination with temozolamide.^78,79

Pharmacokinetic study of intravitreal injections of rituximab in rabbits demonstrated a 4.7-day half-life of the drug in the vitreous.⁸⁰ Extrapolation of these data suggested that a 1-mg intravitreal injection in humans might deliver vitreous drug levels that would remain above 10 ng/mL for 72 days. Drugs are known to be cleared more rapidly in previously vitrectomized eyes, however, and most patients eligible for this therapy have undergone vitrectomy and possibly cataract surgery.⁸⁰ The effect of these surgeries on the clearance of rituximab is not known. Kitzmann and colleagues⁸¹ recently confirmed the lack of toxicity of intravitreal rituximab in rabbits and reported its first use in humans. Five eyes of 3 patients were treated with 1 to 4 injections per eye. There were no signs of clinical toxicity, although no electroretinograms were performed. Clinical effectiveness of the drug could not yet be determined because all 3 patients in the study received either other additional ocular treatment or systemic treatment. More study is necessary to determine the future role of immunotherapy for this disease, but the approach holds promise given the recent success of monoclonal antibody treatments such as anti-VEGF for other ocular diseases such as neovascular AMD.

CONCLUSIONS

Many promising treatments are on the horizon for intraocular tumors. Combination therapeutic regimens for retinoblastoma combining systemic chemotherapy, focal treatment, and antivascular targeting agents hold promise for increased local tumor control in patients with advanced disease. For patients with posterior uveal malignant melanoma, local tumor control with radioactive plaques is typically excellent, but approximately half of patients treated with iodine 125 have poor VA (worse than 20/200) and develop signs of radiation retinopathy within 5 years of treatment. Recent research into alternative adjunctive therapies for saving vision, such as anti-VEGF compounds, is also intriguing. Patients with primary intraocular lymphoma will likely benefit from improved diagnostic testing and immunomodulating therapy in the near future as well. RP

REFERENCES

1. Murphree AL. The Case for a new evidence-based group classification of intraocular retinoblastoma: linking natural history with clinical outcomes. In: International Conference of Ocular Oncology; September, 2005; Whistler, Canada.
2. Shields CL, Shields JA. Basic understanding of current classification and management of retinoblastoma. Curr Opin Ophthalmol. 2006;17:228-234.
3. Abramson DH, Schefler AC. Update on Retinoblastoma. Retina. 2004;24:828-848.
4. Wong FL, Boice JD, Jr., Abramson DH, et al. Cancer incidence after retinoblastoma. Radiation dose and sarcoma risk. JAMA. 1997;278:1262-1267.
5. Murphree AL, Villablanca JG, Deegan WF, 3rd, et al. Chemotherapy plus local treatment in the management of intraocular retinoblastoma. Arch Ophthalmol. 1996;114:1348-1356.
6. Kingston JE, Hungerford JL, Madreperla SA, Plowman PN. Results of combined chemotherapy and radiotherapy for advanced intraocular retinoblastoma. Arch Ophthalmol. 1996;114:1339-1343.
7. Gallie BL, Budning A, DeBoer G, et al. Chemotherapy with focal therapy can cure intraocular retinoblastoma without radiotherapy. Arch Ophthalmol. 1996;114:1321-1328.
8. Shields CL, De Potter P, Himelstein BP, et al. Chemoreduction in the initial management of intraocular retinoblastoma. Arch Ophthalmol. 1996;114:1330-1338.
9. Greenwald MJ, Strauss LC. Treatment of intraocular retinoblastoma with carboplatin and etoposide chemotherapy. Ophthalmology. 1996;103:1989-1997.
10. Shields CL, Shields JA, Needle M, et al. Combined chemoreduction and adjuvant treatment for intraocular retinoblastoma. Ophthalmology. 211997;104:2101-2111.
11. Gunduz K, Shields CL, Shields JA, et al. The outcome of chemoreduction treatment in patients with Reese-Ellsworth group V retinoblastoma. Arch Ophthalmol. 1998;116:1613-1617.
12. Levy C, Doz F, Quintana E, et al. Role of chemotherapy alone or in combination with hyperthermia in the primary treatment of intraocular retinoblastoma: preliminary results. Br J Ophthalmol. 1998;82:1154-1158.
13. Beck MN, Balmer A, Dessing C, et al. First-line chemotherapy with local treatment can prevent external-beam irradiation and enucleation in low-stage intraocular retinoblastoma. J Clin Oncol. 2000;18:2881-2887.
14. Friedman DL, Himelstein B, Shields CL, et al. Chemoreduction and local ophthalmic therapy for intraocular retinoblastoma. J Clin Oncol. 2000;18:12-17.
15. Wilson MW, Rodriguez-Galindo C, Haik BG, et al. Multiagent chemotherapy as neoadjuvant treatment for multifocal intraocular retinoblastoma. Ophthalmology. 2001;108:2106-2114; discussion 14-15.
16. Brichard B, De Bruycker JJ, De Potter P, et al. Combined chemotherapy and local treatment in the management of intraocular retinoblastoma. Med Pediatr Oncol. 2002;38:411-415.
17. Gombos DS, Kelly A, Coen PG, et al. Retinoblastoma treated with primary chemotherapy alone: the significance of tumour size, location, and age. Br J Ophthalmol.2002;86:80-83.
18. Lumbroso L, Doz F, Urbieta M, et al. Chemothermotherapy in the management of retinoblastoma. Ophthalmology. 2002;109:1130-1136.
19. Shields CL, Honavar SG, Meadows AT, et al. Chemoreduction plus focal therapy for retinoblastoma: factors predictive of need for treatment with external beam radiotherapy or enucleation. Am J Ophthalmol. 2002;133:657-664.
20. Sussman DA, Escalona-Benz E, Benz MS, et al. Comparison of retinoblastoma reduction for chemotherapy vs external beam radiotherapy. Arch Ophthalmol. 2003;121:979-984.
21. Rodriguez-Galindo C, Wilson MW, Haik BG, et al. Treatment of intraocular retinoblstoma with vincristine and carboplatin. J Clin Oncol.2003;21:2019-2025.
22. Schueler AO, Jurklies C, Heimann H, et al. Thermochemotherapy in hereditary retinoblastoma. Br J Ophthalmol 2003;87:90-95.
23. Shields CL, Mashayekhi A, Au AK, et al. The International Classification of Retinoblastoma predicts chemoreduction success. Ophthalmology. 2006;113:2276-2280.
24. Schefler AC, Cicciarelli N, Feuer W, et al. Macular retinoblastoma: evaluation of tumor control, local complications, and visual outcomes for eyes treated with chemotherapy and repetitive foveal laser ablation. Ophthalmology. 2007;114:162-169.
25. Dunkel IJ, Lee TC, Shi W, et al. A phase II trial of carboplatin for intraocular retinoblastoma. Pediatr Blood Cancer. 2007.
26. Chan HS, DeBoer G, Thiessen JJ, et al. Combining cyclosporin with chemotherapy controls intraocular retinoblastoma without requiring radiation. Clin Cancer Res. 1996;2:1499-1508.
27. Murray TG, Cicciarelli N, McCabe CM, et al. In vitro efficacy of carboplatin and hyperthermia in a murine retinoblastoma cell line. Invest Ophthalmol Vis Sci. 1997;38:2516-2522.
28. Abramson DH, Gombos DS. The topography of bilateral retinoblastoma lesions. Retina. 1996;16:232-239.
29. Harbour JW, Murray TG, Hamasaki D, et al. Local carboplatin therapy in transgenic murine retinoblastoma. Invest Ophthalmol Vis Sci. 1996;37:1892-1898.
30. Murray TG, Cicciarelli N, O�Brien JM, et al. Subconjunctival carboplatin therapy and cryotherapy in the treatment of transgenic murine retinoblastoma. Arch Ophthalmol. 1997;115:1286-1290.
31. Abramson DH, Frank CM, Dunkel IJ. A phase I/II study of subconjunctival carboplatin for intraocular retinoblastoma. Ophthalmology. 1999;106:1947-1950.
32. Mulvihill A, Budning A, Jay V, et al. Ocular motility changes after subtenon carboplatin chemotherapy for retinoblastoma. Arch Ophthalmol. 2003;121:1120-1124.
33. Burnier MN, McLean IW, Zimmerman LE, Rosenberg SH. Retinoblastoma. The relationship of proliferating cells to blood vessels. Invest Ophthalmol Vis Sci. 1990;31:2037-2040.
34. Schefler AC, Jockovich ME, Toledano S, Murray TG. Historical and modern approaches to chemotherapy. Exp Rev Ophthalmol. 2006;1:83-95.
35. Escalona-Benz E, Jockovich ME, Murray TG, et al. Combretastatin A-4 prodrug in the treatment of a murine model of retinoblastoma. Invest Ophthalmol Vis Sci. 2005;46:8-11.
36. Jockovich ME, Murray TG, Escalona-Benz E, et al. Anecortave acetate as single and adjuvant therapy in the treatment of retinal tumors of LH(BETA)T(AG) mice. Invest Ophthalmol Vis Sci. 2006;47:1264-1268.
37. Jockovich ME, Bajenaru ML, Pina Y, et al. Retinoblastoma tumor vessel maturation impacts efficacy of vessel targeting in the LH(BETA)T(AG) mouse model. Invest Ophthalmol Vis Sci. 2007;48:2476-2482.
38. Yamane T, Kaneko A, Mohri M. The technique of ophthalmic arterial infusion therapy for patients with intraocular retinoblastoma. Int J Clin Oncol. 2004;9:69-73.
39. Abramson DH, Dunkel IJ, Kim J, et al. Direct (Intra) ophthalmic artery delivery of chemotherapy (melphalan) for advanced intraocular retinoblastoma with seeding: alternative to enucleation. In: International Society of Ocular Oncology; June, 2007; Siena, Italy.
40. Jockovich ME, Murray TG, Clifford PD, Moshfeghi AA. Posterior juxtascleral injection of anecortave acetate: magnetic resonance and echographic imaging and localization in rabbit eyes. Retina. 2007;27:247-252.
41. Ruggeri M, Wehbe H, Jiao S, et al. In vivo three-dimensional high-resolution imaging of rodent retina with spectral-domain optical coherence tomography. Invest Ophthalmol Vis Sci. 2007;48:1808-1814.
42. Sisley K, Rennie IG, Cottam DW, et al. Cytogenetic findings in six posterior uveal melanomas: involvement of chromosomes 3, 6, and 8. Genes Chromosomes Cancer. 1990;2:205-209.
43. Prescher G, Bornfeld N, Hirche H, et al. Prognostic implications of monosomy 3 in uveal melanoma. Lancet. 1996;347:1222-1225.
44. Onken MD, Worley LA, Ehlers JP, Harbour JW. Gene expression profiling in uveal melanoma reveals two molecular classes and predicts metastatic death. Cancer Res. 2004;64:7205-7209.
45. Naus NC, Verhoeven AC, van Drunen E, et al. Detection of genetic prognostic markers in uveal melanoma biopsies using fluorescence in situ hybridization. Clin Cancer Res. 2002;8:534-539.
46. Shields CL, Ganguly A, Materin M, et al. Chromosome 3 analysis of uveal melanoma using fine-needle aspiration biopsy at the time of plaque radiotherapy in 140 consecutive cases. Arch Ophthalmol. 2007;125:1017-1024.
47. Mortality in patients with small choroidal melanoma. COMS report no. 4. The Collaborative Ocular Melanoma Study Group. Arch Ophthalmol. 1997;115:886-893.
48. Augsburger JJ. Is observation really appropriate for small choroidal melanomas. Trans Am Ophthalmol Soc. 1993;91:147-68; discussion 69-75.
49. Sobrin L, Schiffman JC, Markoe AM, Murray TG. Outcomes of iodine 125 plaque radiotherapy after initial observation of suspected small choroidal melanomas: a pilot study. Ophthalmology. 2005;112:1777-1783.
50. Butler P, Char DH, Zarbin M, Kroll S. Natural history of indeterminate pigmented choroidal tumors. Ophthalmology. 1994;101:710-716; discussion 7.
51. Singh A, Bena J, Janku L, Schachat A. Prompt vs. deferred treatment of choroidal indeterminate melanocytic lesions: a multicenter randomized trial. In: International Society of Ocular Oncology; June, 2007; Siena, Italy.
52. Gunduz K, Shields CL, Shields JA, et al. Radiation retinopathy following plaque radiotherapy for posterior uveal melanoma. Arch Ophthalmol. 1999;117:609-614.
53. Augsburger JJ, Roth SE, Magargal LE, Shields JA. Panretinal photocoagulation for radiation-induced ocular ischemia. Ophthalmic Surg. 1987;18:589-593.
54. Finger PT, Kurli M. Laser photocoagulation for radiation retinopathy after ophthalmic plaque radiation therapy. Br J Ophthalmol. 2005;89:730-738.
55. Shields CL, Demirci H, Dai V, et al. Intravitreal triamcinolone acetonide for radiation maculopathy after plaque radiotherapy for choroidal melanoma. Retina. 2005;25:868-874.
56. Mashayekhi A, Shields CL, Phan L, et al. Bevcizumab for treatment of cystoid macular edema following plaque radiotherapy of posterior uveal melanoma. In: International Society of Ocular Oncology; June 2007; Siena, Italy.
57. Pilotto E, Vujosevic S, Parrozzani R, et al. Bevacizumab after brachytherapy for choroidal melanoma. In: International Society of Ocular Oncology; 2007; Siena, Italy.
58. Finger PT, Chin K. Anti-vascular endothelial growth factor bevacizumab (avastin) for radiation retinopathy. Arch Ophthalmol. 2007;125:751-756.
59. Becerra E, Kenawy N, Groenewald C, Damato B. Intravitreal bevacizumab (Avastin) in radiation-induced maculopathy. In: International Society of Ocular Oncology; 2007; Siena, Italy.
60. Hui JI, Murray TG. Radioactive plaque therapy. Int Ophthalmol Clin. 2006;46:51-68.
61. Shields CL, Cater J, Shields JA, et al. Combination of clinical factors predictive of growth of small choroidal melanoctic tumors. Arch Ophthalmol. 2000;118:360-364.
62. Lavinsky D, Belfort RN, Navajas EV, et al. Fundus autofluorescence of choroidal nevus and melanoma. Br J Ophthalmol. 2007;91:1299-1302.
63. Factors predictive of growth and treatment of small choroidal melanoma: COMS Report No. 5. The Collaborative Ocular Melanoma Study Group. Arch Ophthalmol. 1997;115:1537-1544.
64. Bardenstein DS. Intraocular Lymphoma. Cancer Control 1998;5:317-325.
65. Melson MR, Mukai S. Intraocular lymphoma. Int Ophthalmol Clin. 2006;46:69-77.
66. Malumbres R, Davis J, Ruiz P, Lossos IS. Somatically mutated immuniglobulin IGHV genes without intraclonal heterogeneity indicate a postgerminal centre origin of primary intraocular diffuse large B-cell lymphomas. British J Hematol. 2007;138:749-755.
67. Bessell EM, Hoang-Xuan K, Ferreri AJ, Reni M. Primary central nervous system lymphoma: biological aspects and controversies in management. Eur J Cancer. 2007;43:1141-1152.
68. Kadan-Lottick NS, Skluzacek MC, Gurney JG. Decreasing incidence rates of primary central nervous system lymphoma. Cancer. 2002;95:193-202.
69. Rothova A, Ooijman F, Kerkhoff F, et al. Uveitis masquerade syndromes. Ophthalmology. 2001;108:386-399.
70. Cassoux N, Giron A, Bodaghi B, et al. IL-10 measurement in aqueous humor for screening patients with suspicion of primary intraocular lymphoma. Invest Ophthalmol Vis Sci. 2007;48:3253-3259.
71. Choi JY, Kafkala C, Foster CS. Primary intraocular lymphoma: a review. Semin Ophthalmol. 2006;21:125-133.
72. Ferreri AJ, Blay JY, Reni M, et al. Relevance of intraocular involvement in the management of primary central nervous system lymphomas. Ann Oncol. 2002;13:531-538.
73. Soussain C, Suzan F, Hoang-Xuan K, et al. Results of intensive chemotherapy followed by hematopoietic stem-cell rescue in 22 patients with refractory or recurrent primary CNS lymphoma or intraocular lymphoma. J Clin Oncol. 2001;19:742-749.
74. Fishburne BC, Wilson DJ, Rosenbaum JT, Neuwelt EA. Intravitreal methotrexate as an adjunctive treatment of intraocular lymphoma. Arch Ophthalmol. 1997;115:1152-1156.
75. de Smet MD, Vancs VS, Kohler D, et al. Intravitreal chemotherapy for the treatment of recurrent intraocular lymphoma. Br J Ophthalmol. 1999;83:448-451.
76. Smith JR, Rosenbaum JT, Wilson DJ, et al. Role of intravitreal methotrexate in the management of primary central nervous system lymphoma with ocular involvement. Ophthalmology. 2002;109:1709-1716.
77. Schulz H, Pels H, Schmidt-Wolf I, et al. Intraventricular treatment of relapsed central nervous system lymphoma with the anti-CD20 antibody rituximab. Haematologica. 2004;89:753-754.
78. Wong ET, Tishler R, Barron L, Wu JK. Immunochemotherapy with rituximab and temozolomide for central nervous system lymphomas. Cancer. 2004; 101:139-145.
79. Enting RH, Demopoulos A, DeAngelis LM, Abrey LE. Salvage therapy for primary CNS lymphoma with a combination of rituximab and temozolomide. Neurology. 2004;63:901-903.
80. Kim H, Csaky KG, Chan CC, et al. The pharmacokinetics of rituximab following an intravitreal injection. Exp Eye Res. 2006;82:760-766.
81. Kitzmann AS, Pulido JS, Mohney BG, et al. Intraocular use of rituximab. Eye. 2007 Apr 20; [Epub ahead of print].
82. Shields CL, Meadows AT, Leahey AM, Shields JA. Continuing challenges in the management of retinoblastoma with chemotherapy. Retina. 2004;24:849-862.